A new meta-heuristic algorithm: Artificial Yellow Ground Squirrel (YGSA)

Abstract

AbstractRecently, extensive research has been focused on population-based and nature-inspired optimization algorithms. Such as war strategy algorithm, particle swarms algorithm, gray wolves algorithm, and other algorithms. Depending on their nature, each algorithm has various applications in different sciences. Despite their benefits, there are a few problems such as convergence and avoid from the trap of local optimum. In this paper, a novel optimization algorithm called Yellow Ground Squirrel Algorithm (YGSA) has been proposed, which has been inspired based on observation of the yellow ground squirrel's behavior. The proposed strategy has been modeled basis of escaping of squirrel and chasing hunter, where the squirrel tries to increase its distance to hunter and to reduce its distance to nest. Squirrel attempts to keep it constant or increasing its distance to hunter to find its next position. The experiments has been evaluated by the 56 benchmark test functions and compared with other meta-heuristic algorithms including HBO, GSA, PSO, SCA, and WSO. The experiment results has demonstrated performance of YGSA in terms of the Convergence, global and local optimal is yield better outcomes against other mentioned meta-heuristic algorithms.